Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Scientific Question Answering on GPQA Main (test)

23.04P@1

DecepChain

2.46887.809413.1518.4906Sep 30, 2025
Updated 12d ago

Evaluation Results

MethodLinks
2025.09
23.0485.0234.98
2025.09
18.3583.218.52
2025.09
17.99--
2025.09
17.7661.470
2025.09
17.63--
2025.09
16.7--
2025.09
16.07--
2025.09
15.6787.1417.93
2025.09
15.2588.4824.41
2025.09
14.8781.880
2025.09
14.7888.824.23
2025.09
13.4886.160
2025.09
12.3782.510
2025.09
11.3487.460
2025.09
11.2186.860
2025.09
8.381.430
2025.09
7.3778.840
2025.09
4.5174.420
2025.09
3.3994.260
2025.09
3.2693.530